Ooodles of old stuff including How-To's, Writers Resources, Directories of all Types, Technology Reviews, Health Information, Marketing Statisitcs, Affluent Markets, Hospital and Medical Market Data and more

Monday, September 04, 2006

Statistics in Life 9

Okay, for the entry. I've been reading a few articles and listening to a guest speaker on evaluation for another course. What struck me as strange (maybe this is obvious) is that research and evaluation are not the same thing. I always thought they were. According to this speaker, evaluation measures the degree of effectiveness whereas research (as mentioned in this class) measures the probability of a mistake/level of significance. Research may or may not have application in "real life" but evaluation is always measured in terms of "real-life". There is a null hypothesis and an alternative hypothesis in research, but in evaluation…depending upon the criteria, "no-change" may be considered enough for a positive evaluation. As if I wasn't confused enough, I fail to understand the difference between how to tell the difference between evaluating something where "no change" is a "good" evaluation and instead measuring it against the normally expected deterioration. For example, if I were testing an intervention … say an elderly persons ability to remain at home and function independently…then the evaluation would consist of meeting the criteria of staying at home. Now when I went to measure that for evaluation purposes, the goal was met with no change resulting. Research would fail to show anything significant having occurred. I realize research demonstrates scientific knowledge and the intervention failed to give results that can be "translated" into the scientific body of knowledge. But, did it fail to give generalizable or valid results simply because it lacked statistical support? For research to benefit I would have to design a study to comparing mean rates of independent living to the mean rate of others without this intervention who then end up in a nursing home to find if it is significant.


Post a Comment

Links to this post:

Create a Link

<< Home